Radical-mediated dehydrogenation of bile acids by means of hydrogen atom transfer to triplet carbonyls

Article abstract of DOI:10.1039/c5ob02561c

The aim of the present paper is to explore the potential of radical-mediated dehydrogenation of bile salts (BSs), which is reminiscent of the enzymatic action of hydroxysteroid dehydrogenase enzymes (HSDH). The concept has been demonstrated using triplet carbonyls that can be efficiently generated upon selective UVA-excitation. Hydrogen atom transfer (HAT) from BSs to triplet benzophenone (BP) derivatives gave rise to radicals, ultimately leading to reduction of the BP chromophore with concomitant formation of the oxo-analogs of the corresponding BSs. The direct reactivity of triplet BP with BSs in the initial step was evaluated by determining the kinetic rate constants using laser flash photolysis (LFP). The BP triplet decay was monitored (λ_max = 520 nm) upon addition of increasing BS concentrations, and the obtained rate constant values indicated a reactivity of the methine hydrogen atoms in the order of C-3 < C-12 < C-7. The steady-state kinetics of the overall process, monitored through the disappearance of the typical BP absorption band at 260 nm, was much faster under N₂ than under O₂, also supporting the role of the oxygen-quenchable triplet in the dehydrogenation process. Furthermore, irradiation of deaerated aqueous solutions of sodium cholate in the presence of KPMe provided the oxo-analogs, 3[O],7[O]-CA, 3[O]-CA and 7[O]-CA, arising from the HAT process.

Full text of DOI:10.1039/c5ob02561c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Radical-mediated dehydrogenation of bile acids by
means of hydrogen atom transfer to triplet
carbonyls†
Cite this: Org. Biomol. Chem., 2016,
14, 2679
P. Miro, M. L. Marin* and M. A. Miranda*
The aim of the present paper is to explore the potential of radical-mediated dehydrogenation of bile salts
(BSs), which is reminiscent of the enzymatic action of hydroxysteroid dehydrogenase enzymes (HSDH).
The concept has been demonstrated using triplet carbonyls that can be efficiently generated upon selec-
tive UVA-excitation. Hydrogen atom transfer (HAT) from BSs to triplet benzophenone (BP) derivatives gave
rise to radicals, ultimately leading to reduction of the BP chromophore with concomitant formation of the
oxo-analogs of the corresponding BSs. The direct reactivity of triplet BP with BSs in the initial step was
evaluated by determining the kinetic rate constants using laser flash photolysis (LFP). The BP triplet decay
was monitored (λmax = 520 nm) upon addition of increasing BS concentrations, and the obtained rate
constant values indicated a reactivity of the methine hydrogen atoms in the order of C-3 < C-12 < C-7.
The steady-state kinetics of the overall process, monitored through the disappearance of the typical BP
absorption band at 260 nm, was much faster under N2 than under O2, also supporting the role of the
oxygen-quenchable triplet in the dehydrogenation process. Furthermore, irradiation of deaerated
aqueous solutions of sodium cholate in the presence of KPMe provided the oxo-analogs, 3[O],7[O]-CA,
3[O]-CA and 7[O]-CA, arising from the HAT process.
Received 14th December 2015,
Accepted 21st January 2016
DOI: 10.1039/c5ob02561c
www.rsc.org/obc
Introduction
Bile acids (BAs) are a family of amphiphilic steroids that play a
pivotal role in a wide variety of physiological functions, such
as elimination of cholesterol or solubilization of lipids.1–3
Chemically they share a steroidal skeleton with an unusual cis
fusion between rings A and B, a short lateral chain ending in a
carboxylic moiety and different numbers of hydroxyl groups on
the α-face. These structural characteristics contribute to their
facial amphiphilia, associated with a convex hydrophobic
α-face and a concave and hydrophilic β-face (Fig. 1).4 Primary
BAs, cholic acid (CA) and chenodeoxycholic acid (CDCA), are
biosynthesized directly from cholesterol in the hepatocytes of
the liver; secondary BAs, deoxycholic acid (DCA) and litho-
cholic acid (LA), are produced in the colon by enzymatic deoxy-
genation of the hydroxyl group at the C-7 position of CA and
CDCA respectively. Ursodeoxycholic acid (UDCA) is the epimer
of CDCA at C-7 and is also found at low concentrations in
humans.5,6 The amphiphilic properties of BAs allow aggrega-
Fig. 1 Chemical structure of bile acids.
tion in aqueous solution by the interaction of the β-faces of
the monomers, giving rise to the primary aggregates, that
agglomerate as the concentration increases.7–9
Bacterial transformations of BAs have recently attracted
considerable interest in connection with diseases of the
human gastrointestinal tract and, more specifically, colon
cancer.10 In addition, products obtained in the biotransform-
ation of BAs by bacteria have been exploited in the commercial
production of biologically active compounds.11
Instituto Universitario Mixto de Tecnología Química (UPV-CSIC), Departamento de
Química. Universitat Politècnica de València, Avda. los Naranjos s/n 46022, Valencia,
Spain. E-mail: mmiranda@qim.upv.es, marmarin@qim.upv.es
†Electronic supplementary information (ESI) available: Synthesis, exact mass
and spectroscopic characterization (1H, 13C and DEPT-NMR, NOEDIFF) of all
compounds. See DOI: 10.1039/c5ob02561c
The microbial transformations of BAs include dehydrogena-
tion of the α-OH groups at positions C-3, C-7 and/or C-12,
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem., 2016, 14, 2679–2683 | 2679

View Article Online
Paper
Organic & Biomolecular Chemistry
which results in the oxidation to the corresponding oxo-deriva-
tives in a reversible, regio and stereoselective manner.12 In
Results and discussion
general, these processes are driven by hydroxysteroid dehydro- A simplified picture of the radical-mediated approach to
genases (HSDHs) that belong to the family of oxidoreductase achieve dehydrogenation of BSs is shown in Fig. 4. It involves
type enzymes and use the nicotinamide adenine cofactors generation of free radicals through hydrogen atom transfer
NADPH/NADP+ or NADH/NAD+ with an efficient recycling (HAT) from BSs to triplet BP derivatives, ultimately leading to
system,13 such as the pyruvate/lactate pair (Fig. 2).
reduction of the BP chromophore concomitantly with
The potential of HSDHs as a tool to discover new lead com- formation of the oxo-analogs of the corresponding BSs.
pounds for the pharmaceutical industry has been explored by Accordingly, the direct reactivity of triplet BP with BSs in the
means of combinatorial biocatalyzed approaches. This strategy initial step was evaluated by determining the kinetic rate con-
has been effectively used to generate libraries of BA derivatives stants using laser flash photolysis (LFP), and the overall reac-
suitable for high-throughput biological screening.11 An interest- tion was monitored through the disappearance of the typical
ing application is the chemoenzymatic/multienzymatic synthesis BP absorption band at 260 nm. The outcome of the reaction
of UDCA,14,15 a drug used to dissolve cholesterol gallstones.16
was assessed by isolation and full spectroscopic characteri-
With this background, the aim of the work described in the zation of the dehydrogenated BS products.
present paper is to explore the potential of radical-mediated
Generation and detection of free radicals
dehydrogenation of bile salts (BSs), which would be reminis-
cent of the enzymatic action of HSDH. As this new approach is The generation and behavior of the triplet excited state of the BP
based on a different reaction mechanism, it could allow, after chromophore in the presence of BSs was investigated by LFP.
optimization, the achievement of complementary regio and Thus, upon laser excitation (λ = 266 nm) of ketoprofen methyl
stereoselectivities that could expand the available synthetic ester (KPMe) in a deaerated saline aqueous solution ([NaCl] =
toolbox. To prove the concept, carbonyl triplets, such as those 0.2 M), a transient absorption spectrum with two bands peaking
of benzophenone (BP) derivatives (Fig. 3), have been chosen at 330 and 520 nm, characteristic of the triplet excited state
because they can be efficiently generated upon selective UVA- (3KPMe*), was observed (Fig. 5 top). However, when the spectrum
excitation and have reactive nπ* electronic configuration. In was taken in the presence of BSs (NaCA, NaCDCA, NaDCA, NaLA
fact, they are known to act as formal biradicals, mediating or NaUDCA) a new transient signal with two maxima at 340 and
H-abstraction from allylic positions in polyunsaturated fatty 550 nm, characteristic of the benzhydryl radical (KPMe-H•) was
acids17,18 and cholesterol,19,20 a key step in type I lipid observed in all cases as a proof of the HAT process (see Fig. 5
peroxidation.
bottom). The time-evolution of the two species was markedly
In spite of the intensive efforts devoted to developing different, with KPMe-H• decaying much slower than 3KPMe*.
alcohol oxidation methodologies, the photochemical dehydro-
genation for synthetic purposes remains almost unexplored,
Radical-mediated reactivity
3
and the reported examples are essentially restricted to benzyl To quantify the reactivity of KPMe* with BSs, the quenching
alcohols.21
rate constants were determined. Hence, the triplet decay was
Fig. 2 An example of hydroxysteroid dehydrogenase (HSDH)
performance.
Fig. 4 Simplified mechanistic scheme of the radical-mediated dehy-
drogenation of the 3α-position of a BS, triggered by a triplet carbonyl.
Fig. 3 Chemical structures of BP and BP-derivatives.
2680 | Org. Biomol. Chem., 2016, 14, 2679–2683
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 6 Top: kinetic traces recorded at 520 nm after laser flash
irradiation (λexc = 266 nm) of deaerated aqueous solution ([NaCl] = 0.2
M) of KPMe (1 × 10−5 M) in the presence of increasing concentrations of
NaCA (0–17 mM) as an example. Bottom: Stern–Volmer plots of the
Fig. 5 Transient absorption spectra (λexc = 266 nm) obtained for deaer-
ated aqueous solutions ([NaCl] = 0.2 M) of KPMe (1 × 10−5 M) in the
absence (top) or in the presence of NaCA (15 mM, bottom), at different
times after the laser pulse.
●
◆
)
reciprocal KPMe triplet lifetime versus [NaCA]
(
), [NaCDCA]
(
■
○
◊
[NaDCA] ( ), [NaLA] ( ) and [NaUDCA] ( ).
monitored (λmax = 520 nm) upon increasing BSs concentrations
and, in fact, a progressive shortening was observed in all cases
(see Fig. 6 top for NaCA as a representative example). The
reverse of the triplet lifetime was plotted versus BS concen-
trations, and the corresponding rate constant values were
obtained from the slopes of the linear fittings (Fig. 6 bottom
and Table 1). In general, the observed reactivity for the initial
HAT step followed the same trend as the number of CHOH
moieties in the BS, thus indicating that the isopropanol-like
methine hydrogens are indeed the most reactive sites.
Table 1 Determined quenching rate constants for the reaction of
3KPMe* with BSs
BS
kq (M−1 s−1
)
NaCA
3.7 × 108
2.5 × 108
1.4 × 108
4.1 × 107
2.6 × 108
NaCDCA
NaDCA
NaLA
NaUDCA
The regioselectivity of HAT was evaluated by comparing the
quenching rate constants obtained for NaCDCA, NaDCA and
NaUDA, having the same number of CHOH moieties. The
values found for NaCDCA and NaUDCA having the two
hydroxyl groups at C-3 and C-7 (either on the α- or β-face) were
similar; by contrast, a remarkably lower value was found for
NaDCA whose hydroxyl groups were at C-3 and C-12. Assuming
an additive effect and comparing these values to the one
obtained for NaLA, in which only one hydroxyl group (at C-3)
is present, the contribution of each position could be inferred.
Therefore, the order of reactivity of the hydrogen atom in the
HAT step was C-3 < C-12 < C-7. Nonetheless, it has to be con-
sidered that the processes occurring to the long-lived radical
species KPMe-H• are much slower than the initial HAT; hence,
the rate determining step for the overall reaction is not triplet
quenching by the BSs.
O2, supporting the role of the oxygen-quenchable 3KPMe* in
the dehydrogenation process.
Isolation and characterization of dehydrogenated BA
derivatives
Initially, for analytical purposes, an acetonitrile solution of the
methyl ester of CA (CAMe) and BP was irradiated (λirr centered
at 350 nm) under a N2 atmosphere. The resulting crude photo-
mixture was concentrated and purified by column chromato-
graphy yielding three products (Fig. 8). Their structural
1
characterization was based on H and 13C NMR spectroscopy,
including NOEDIFF experiments, combined with exact mass
determination. The first eluted product (A) showed a loss of
four amu in the MS, together with the appearance of two new
13C signals at >210 ppm and the disappearance of two of the
signals in the 65–75 ppm region, thus pointing to the oxi-
dation of two out of the three hydroxyl groups. Moreover, the
second (B) and third (C) eluted products showed a loss of two
The steady-state photolysis of KPMe with NaCA in solution
was monitored by UV-spectrophotometry, following the charac-
teristic KPMe band at 260 nm (Fig. 7). As expected for a triplet
reaction, the kinetics was much faster under N2 than under
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem., 2016, 14, 2679–2683 | 2681

View Article Online
Paper
Organic & Biomolecular Chemistry
fact, when the latter was irradiated, a NOE effect was found on
19-CH3 (0.72 ppm) and 21-CH3 (0.97 ppm, d), while irradiation
of a proton at 3.92 ppm did not result in any NOE effect on the
methyl groups indicating that this signal corresponds to 7β-H.
Finally, compound C was identified as the one resulting from
oxidation at the C-7 position, namely 7[O]-CAMe, based on the
NOE effect observed on the methyl groups 21-CH3 (d) and
19-CH3 (s) upon irradiation on 12β-H (4.02 ppm).
After isolation and unequivocal characterization of the
methyl esters, the three oxo-analogs were subjected to saponi-
fication and used to monitor the kinetics of the photooxida-
tion of NaCA in deaerated aqueous solution ([NaCl] = 0.2 M),
in the presence of KPMe, to mimic the conditions used in the
LFP experiments. The kinetics of the photooxidation was
monitored by UPLC-MS operating in negative ionization mode.
The obtained results (see Fig. 8 as well as Fig. S1 and Tables
S1–S4 in ESI†) indicate that again the main photoproducts
arise from the oxidation at C-3 and/or C-7.
Fig. 7 Relative absorbance of KPMe measured at 260 nm versus
irradiation time (λirr centered at 350 nm) in aqueous solutions ([NaCl] =
0.2 M) containing KPMe (8 × 10−5 M) and NaCA (1 × 10−2 M) under a N2
−5
●
■
) atmosphere. Inset: UV-Vis spectra of KPMe (8 × 10 M)
(
) or an O2
(
and NaCA (1 × 10−2 M) under N2 at different irradiation times.
Although the rate constants determined by LFP could, in
principle, appear inconsistent with the product distributions
obtained in steady-state photolysis, the observed differences
can be explained taking into account the fact that the rate con-
stants are associated with the initial step while the product
distributions are the result of the global process that includes
disproportionation or reversible HAT in the primary radical
pair. Indeed, the long lifetime of KPMe-H• supports that the
subsequent processes occurring from this species are in the
origin of the observed regioselectivity.
Conclusions
Fig. 8 Irradiation (λirr centered at 350 nm) of CAMe or NaCA in the
presence of KPMe in deaerated MeCN or H2O solution, respectively.
Dehydrogenation of bile acids at positions C-3 and/or C-7 has
been achieved by a radical-mediated mechanism, which is remi-
niscent of the enzymatic action of hydroxysteroid dehydrogenases.
The triplet excited states of benzophenone derivatives, generated
upon selective UVA-excitation, have proven to be appropriate
radical precursors, which operate through hydrogen atom transfer
from the methine CHOH groups. As this new approach is based
on a different reaction mechanism, it could allow, after optimi-
zation, achievement of complementary regio and stereo-
selectivities that could expand the available synthetic toolbox.
amu in the MS and the displacement of one 13C signal from
65–75 ppm to >210 ppm; thus, the oxidation of one of the
hydroxyl groups at C-3, C-7 or C-12 was inferred in both cases.
To unequivocally characterize the three photoproducts,
NOEDIFF experiments were performed, and the resulting
spectra were thoroughly analyzed. Compound A was un-
ambiguously assigned as the compound resulting from the oxi-
dation at C-3 and C-7 positions, 3[O],7[O]-CAMe. In fact,
irradiation of proton at 4.06 ppm produced a NOE effect on
methyl groups at 0.98 (d) and 0.72 (s) ppm. This indicates that
the proton at 4.06 ppm corresponds to 12β-H, and the methyl
groups to 21-CH3 and 19-CH3. Furthermore, irradiation of the
methyl group at 1.28 ppm resulted in a NOE effect only on the
signal at 0.72 ppm, indicating that the former corresponds to
18-CH3. Compound B was identified as arising from the oxi-
dation at C-3, namely 3[O]-CAMe, based on the NOE enhance-
ment on methyl groups at 0.97 (s and d) and on the proton at
4.03 ppm upon irradiation of the methyl group at 0.72 ppm.
From this result, the methyl group at 0.72 ppm can be
assigned to 19-CH3 and the singlet at 4.03 ppm to 12β-H. In
Experimental
Irradiation of CAMe with BP
A solution of CAMe (0.557 g, 1.37 mmol) and BP (0.250 g,
1.37 mmol) in CH3CN (150 mL), placed in a pyrex flask was
purged with N2 and irradiated in a photoreactor (λmax
=
350 nm) for six hours. Then, the solvent was concentrated
under vacuum and the crude was purified by column chromato-
graphy (SiO2, AcOEt : Hex, 70 : 30) to yield 3[O],7[O]-CAMe
(0.027 g, 4.8%), 3[O]-CAMe (0.041 g, 7.3%) and 7[O]-CAMe
(0.116 g, 21%). After isolation each photoproduct was
repurified by reverse phase column chromatography
2682 | Org. Biomol. Chem., 2016, 14, 2679–2683
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
before characterization (Li Chroprep RP-18, CH3CN : H2O, Valenciana (Prometeo Program) and the Technical University
60 : 40). of Valencia (Predoctoral FPI fellowship for P. Miró) is gratefully
Characterization of 3[O],7[O]-CAMe. 1H NMR (300 MHz, acknowledged.
CDCl3): δ (ppm) 0.71 (s, 3H, 19-CH3); 0.97 (d, J = 6.0 Hz, 3H,
21-CH3); 1.28 (s, 3H, 18-CH3); 2.86 (dd, J = 13.2 and 4.8 Hz,
1H); 3.66 (s, 3H, CH3O); 4.04 (br s, 1H, 12β-H). 13C NMR
(75 MHz, CDCl3): δ (ppm) 210.8 (C), 210.4 (C), 174.7 (C), Notes and references
72.1 (CH), 51.7 (CH3), 49.7 (CH), 47.8 (CH), 46.8 (CH), 46.7 (C),
1 A. F. Hofmann, Arch. Intern. Med., 1999, 159, 2647–2658.
2 A. F. Hofmann and L. R. Hagey, Cell. Mol. Life Sci., 2008,
65, 2461–2483.
3 A. F. Hofmann, L. R. Hagey and M. D. Krasowski, J. Lipid
Res., 2010, 51, 226–246.
4 A. F. Hofmann and K. J. Mysels, Colloids Surf., 1988, 30,
145–173.
5 A. F. Hofmann, Am. J. Surg., 1989, 158, 198–204.
6 M. J. Armstrong and M. C. Carey, J. Lipid Res., 1982, 23, 70–
80.
7 D. M. Small, S. A. Penkett and D. Chapman, Biochim.
Biophys. Acta, 1969, 176, 178–189.
8 M. Gomez-Mendoza, M. L. Marin and M. A. Miranda,
J. Phys. Chem. Lett., 2011, 2, 782–785.
9 M. Gomez-Mendoza, M. L. Marin and M. A. Miranda,
J. Phys. Chem. B, 2012, 116, 14776–14780.
10 J. M. Ridlon, D.-J. Kang and P. B. Hylemon, J. Lipid Res.,
2006, 47, 241–259.
11 F. Secundo, G. Carrea, M. De Amici, S. Joppolo di Ventimi-
glia and J. S. Dordick, Biotechnol. Bioeng., 2003, 81, 391–
396.
12 V. Prabha and M. Ohri, World J. Microbiol. Biotechnol., 2006,
22, 191–196.
13 A. K. Agarwal and R. J. Auchus, Endocrinology, 2005, 146,
2531–2538.
14 P. P. Giovannini, A. Grandini, D. Perrone, P. Pedrini,
G. Fantin and M. Fogagnolo, Steroids, 2008, 73, 1385–
1390.
45.1 (CH2), 43.1 (CH2), 40.8 (CH), 36.8 (CH2), 36.2 (CH),
35.5 (CH2), 35.0 (C + CH), 31.2 (CH2), 31.0 (CH2), 29.7 (CH2),
27.7 (CH2), 24.4 (CH2), 22.4 (CH3), 17.7 (CH3), 13.1
(CH3). m/z found 419.2813, calculated for C25H39O5 (MH+)
419.2797.
Characterization of 3[O]-CAMe. 1H NMR (300 MHz, CDCl3):
δ (ppm) 0.72 (s, 3H, 19-CH3); 0.97 (m, 6H, 21-CH3 + 18-CH3);
3.39 (dd, J = 15.0 and 13.5 Hz, 1H); 3.66 (s, 3H, CH3O); 3.91 (br
s, 1H, 7β-H); 4.02 (br s, 1H, 12β-H). 13C NMR (75 MHz, CDCl3):
δ (ppm) 213.3 (C), 174.9 (C), 73.0 (CH), 68.4 (CH), 51.7 (CH3),
47.4 (CH), 46.8 (C), 45.7 (CH2), 43.2 (CH), 42.0 (CH), 39.7 (CH),
37.0 (CH2), 36.8 (CH2), 35.3 (CH), 35.0 (C), 34.0 (CH2), 31.2
(CH2), 31.0 (CH2), 28.8 (CH2), 27.6 (CH2), 27.4 (CH), 23.3
(CH2), 21.8 (CH3), 17.5 (CH3), 12.7 (CH3). m/z found 443.2756,
calculated for C25H40O5Na (MNa+) 443.2773.
Characterization of 7[O]-CAMe. 1H NMR (300 MHz, CDCl3):
δ (ppm) 0.68 (s, 3H, 19-CH3); 0.97 (d, J = 6.0 Hz, 3H, 21-CH3);
1.18 (s, 3H, 18-CH3); 2.83 (dd, J = 12.4 and 6.4 Hz, 1H); 3.58
(m, 1H, 3β-H); 3.66 (s, 3H, CH3O); 4.01 (br s, 1H, 12β-H).
13C NMR (75 MHz, CDCl3): δ (ppm) 211.5 (C), 174.7 (C),
72.2 (CH), 71.1 (CH), 51.7 (CH3), 49.7 (CH), 46.7 (CH + C),
46.2 (CH), 45.5 (CH2), 40.9 (CH), 37.6 (CH2), 36.1 (CH),
35.1 (CH), 34.8 (C), 34.2 (CH2), 31.2 (CH2), 31.0 (CH2),
30.0 (CH2), 29.4 (CH2), 27.7 (CH2), 24.4 (CH2), 23.0 (CH3),
17.6 (CH3), 13.0 (CH3). m/z found 421.2962, calculated for
C25H41O5 (MH+) 421.2954.
Irradiation of NaCA with KPMe
15 D. Monti, E. E. Ferrandi, I. Zanellato, L. Hua, F. Polentini,
G. Carrea and S. Riva, Adv. Synth. Catal., 2009, 351, 1303–
1311.
16 G. Salen, A. Colalillo, D. Verga, E. Bagan, G. S. Tint and
S. Shefer, Gastroenterology, 1980, 78, 1412–1418.
17 M. V. Encinas and J. C. Scaiano, J. Am. Chem. Soc., 1981,
103, 6393–6397.
A solution of NaCA (0.215 g, 0.5 mmol) and KPMe (1.86 g,
0.5 mmol) in 100 mL of aqueous solution ([NaCl] = 0.2 M),
placed in a pyrex flask, was purged with N2 and irradiated in
a photoreactor (λmax = 350 nm) for twelve hours. The kinetics
of the photoreaction was analyzed at different times.
The photoproducts in the crude were identified by
comparison with isolated standards obtained upon saponifica-
tion of 3[O],7[O]-CAMe, 3[O]-CAMe and 7[O]-CAMe in
aqueous NaOH, 1 M. Quantification was based on area
integration.
18 D. Z. Markovic, T. Durand and L. K. Patterson, Photochem.
Photobiol., 1990, 51, 389–394.
19 I. Andreu, I. M. Morera, F. Bosca, L. Sanchez, P. Camps and
M. A. Miranda, Org. Biomol. Chem., 2008, 6, 860–867.
20 D. Neshchadin, F. Palumbo, M. S. Sinicropi, I. Andreu,
G. Gescheidt and M. A. Miranda, Chem. Sci., 2013, 4, 1608–
1614.
Acknowledgements
Financial support from the Spanish Government (Grants 21 S. Kamijo, K. Tao, G. Takao, H. Tonoda and T. Murafuji,
SEV-2012-0267 and CTQ2012-38754-C03-03), Generalitat
Org. Lett., 2015, 17, 3326–3329.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem., 2016, 14, 2679–2683 | 2683